Mark Towers
Software Engineer at Anyscale
London, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolMark Towers is a software engineer with a decade of experience focused on reinforcement learning infrastructure and tooling, currently developing RLlib at Anyscale. He holds a PhD in Computer Science from the University of Southampton where he researched explainable reinforcement learning, and previously built LLM-powered automation for finance during a data science internship. An active open-source maintainer and contributor, Mark helps steer Gymnasium and Farama-Foundation projects like PettingZoo and Minigrid, improving cross-environment robustness, test quality, and API consistency. His work spans backend systems, test automation, and environment design, with a pragmatic eye for reducing warnings, fixing subtle RNG bugs, and modernizing build processes. Based in London, he combines research-grade rigor with production-focused engineering, often surfacing small fixes that prevent brittle behavior across many RL environments.
10 years of coding experience
1 year of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of Southampton
Github Skills (20)
git10
environmental10
debug10
dev-environment10
pytest10
multi-agent-reinforcement-learning10
python10
gymnasium10
openai-gym10
enviroment10
numpy10
api10
environ10
reinforcement-learning9
refactor9
Programming languages (12)
Github contributions (5)
Farama-Foundation/Gymnasium
Sep 2022 - Jan 2023
An API standard for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)
Role in this project:
Back-end Developer & Test Automation Engineer
Back-end Developer & Test Automation EngineerContributions:15 releases, 914 reviews, 188 commits in 4 months
Contributions summary:Mark's primary contribution involved renaming the core project from "Gym" to "Gymnasium" and adapting the codebase to reflect this change. This included modifications across multiple test files, demonstrating a focus on ensuring the codebase continued to function as expected after the renaming. The user also updated the build process, including changes to the Dockerfile, to reflect the new project name.
apiagentreinforcement-learningreinforcement-learning-agentgymnasium
openai/gym
Mar 2022 - Oct 2022
A toolkit for developing and comparing reinforcement learning algorithms.
Role in this project:Back-end Developer & QA Engineer
Back-end Developer & QA EngineerContributions:387 reviews, 64 commits, 82 PRs in 6 months
Contributions summary:Mark primarily contributed to improving the quality and robustness of the OpenAI Gym toolkit. They focused on reducing warnings produced by pytest and addressed several issues, including incorrect data types and unclosed video recorders. The user implemented fixes related to environment rendering and ensured proper rendering across environments while also providing better parametrization for easier debugging of tests.
comparingreinforcement-learning-algorithmsdevelopingdeep-learningreinforcement-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.